Introduction: Csikszentmihalyi (1975) proposed flow as a nine dimensional construct that reflects optimal experiences in various activities. In a recent review Csikszentmihalyi (2000) advocated that some flow factors are common conditions for flow experiences, such as challenge-skills balance, clear goals, and unambiguous feedback, whereas the other flow factors could be categorised as characteristics of flow. Previous research corroborated these propositions, indicating that flow factors of action-awareness merging, loss of self-consciousness, and time transformation had a weaker connection with flow (Jackson, Kimiecik, Ford, & Marsh, 1998; Jackson, Thomas, Marsh, & Smethurst, 2001). Following theoretical propositions by Csikszentmihalyi (2000) and previous research findings, the aim of this study was the examination of two flow models using a nine or a six-factor structure. Methods Sample The sample (N = 271) consisted of 187 male and 84 female players, between 11 and 18 years of age (M = 14.31; SD = 1.59), who participated frequently in tennis competitions. Participants had been involved in tennis for an average of 6.66 years (SD = 2.51) and in tennis competitions for a mean of 4.28 years (SD = 1.88). Participants had a mean training intensity of 9.19 hours per week (SD = 5.87). Measures The Dispositional Flow Scale-2 (DFS-2) consists of 36 items representing nine subscales, each comprising four items assessing one of the nine dimensions of flow. The subscales showed acceptable reliability values, ranging between .81 and .90. The response format is a 5-point Likert scale anchored by 1 (never) and 5 (always). Procedure The research was approved by the Victoria University Ethics Committee. We requested access from a range of tennis centres and tournament directors in metropolitan Melbourne, Australia. Participants completed the measures between training sessions or competition matches. Statistical Analysis Structural equation modelling was used to test the exogenous variables of action control, imagery use, absorption, and trait sport confidence with the endogenous variable dispositional flow. The analysis was run twice with a nine- and a six-factor model of dispositional flow. Results Using structural equation modelling, the nine-factor flow model showed a significant chi-square of 2 (N = 271, df = 59) = 133.13, p < .01. The minimum sample discrepancy divided by degrees of freedom (CMIN/DF) showed an acceptable score of 2.26 and the Root Mean Square Error of Approximation (RMSEA) was .07. The Normed Fit Index (NFI) showed a score of .88, the Goodness of Fit Index (GFI ) was .93, and the Comparative Fit Index (CFI) was .93. The model fit summary for the six-factor model showed 2 (N = 271, df = 29) = 62.51, p < .01. The model summary revealed a slightly better fit for CMIN/DF (2.16), RMSEA (.07), NFI (.94), GFI (.95), and CFI (.96). Discussion The results indicated that the six-factor model was a better representation of dispositional flow than the original nine-factor solution. These findings corroborate previous research (Jackson et al., 1998, 2001) that found factors of action-awareness merging, loss of self-consciousness, and time transformation contributed less to the dispositional flow than the remaining factors. More research is needed to substantiate a six-factor or nine-factor model of flow to facilitate a differentiated view of flow dimensions in sports.
© Copyright 2008 2008 International Convention on Science, Education and Medicine in Sport: Proceedings, Vol. III. Veröffentlicht von People´s Sports Publishing House. Alle Rechte vorbehalten.